Reading 35, Brinson model

so it looks like there are two formulas for allocation effect, and most questions i came across required to use the second formula. my question is when to use the first formula?
Ai = (wi – Wi)Bi
Ai = (wi – Wi)(Bi – B)

I should think that they would have to specify which model they want you to use:

  • Brinson, for which Ai = (wiWi)Bi
  • Brinson-Fachler, for which Ai = (wiWi)(BiB)

Both of these formula are OK however please note the shorter of the two can only be used when you have the whole of the portfolio allocation ie the sum of the weights is 100%. If instead you are only given information on say one sector then you must use the longer formula. The longer formula is therefore the safer one to use since it works in all instances.

If I’m not mistaken
Bronson-fletcher sometimes can be asked as
1- Macro Effect => use B-F model
2- Allocation/Board/Investment Committee/ Top down effect: => use B-F model
The reason for number 2 if I remember correctly that the portfolio will have limited decision on Asset Allocation and where he can add value is between security selection…
Again I might be wrong but that’s what I recall, I had trouble understanding it until I had it drawn as a kind map in a piece of paper :stuck_out_tongue: all the best

Except that they’d spell it correctly: Brinson-Fachler.

Ops

:wink:

you are right!

Hi all,

I having the same doubt as well and looking through at the thread. It seems that unless the CFAI state explicitly indicate the model to use (be it Brinson model or Brinson-Fachler model, both approaches calculation are not wrong per se.

@Simon1, I hope your advice works but it does not seem to apply. In the CFAI curriculum, under Reading 35,exhibit 12 page 200, the exhibit shows effect of industry sector allocation with portfolio weight adding up to 100% and the allocation and selection calculation for energy, health care, financial sector etc using BF model aka the “longer” formula.

My question is for some of attempted mock exams, some of the providers have asked to calculate selection effect /allocation effect without indicating the model explicitly. It is only when looking at the guideline answers we are able to determine the model used. So how so we know which model to use if the mock questions did not explicitly indicate the model used and how do we know that the guideline answers are corrected themselves?

Am I missing something?

Hi, apologies for not being clearer, but what I meant is there is a difference between performance attribution for the whole portfolio (when we use the simpler formula) versus attribution for a specific sector or part of the portfolio, then you need to use the longer formula.

No.

It’s a matter of which model you choose to use, not whether you’re looking at the whole portfolio or only part of it.

Hi Magician,
Thank you. My question is since it depend on the model used, will it be stated explicitly during the actual exam and as such, we will just need to rote memorisation the formulas for Brinson and BF model?

Also, in the some old past year CFAI exams, some of the questions use of words as such “selection(including interaction), assuming selection-only decision and /or sector-allocation decision” etc, are we suppose to know these meaning as well for current 2021curriculum as I gone through theCFAI offical curriculum but can’t seem to find them though. Or they are no longer relevant?